{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "53c733c8-68f5-4348-92f7-39d3223cc7c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_3137448/2469479857.py:12: DtypeWarning: Columns (2) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  df = pd.read_csv('/home/ozturan/covid-conspiracy-narratives/src/musk/submission/figures/deca_mis_conc_with_shortener3.tsv', sep= '\\t')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "\n",
    "newsguard = pd.read_csv('/net/lazer/lab-lazer/shared_projects/datasets/newsguard/metadata/2023-05.csv')\n",
    "newsguard.rename(columns={\"Domain\": \"expanded_domain\"}, inplace=True)\n",
    "newsguard = newsguard[~newsguard['Rating'].isin(['P', 'S'])]\n",
    "newsguard = newsguard.drop_duplicates(subset=['expanded_domain'], keep='first').reset_index(drop=True)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "df = pd.read_csv('/home/ozturan/covid-conspiracy-narratives/src/musk/submission/figures/deca_mis_conc_with_shortener3.tsv', sep= '\\t')\n",
    "domain_ratings = pd.read_csv('/home/ozturan/covid-conspiracy-narratives/src/domain_ratings_with_shorteners.csv')\n",
    "domain_ratings  = domain_ratings.drop(columns=['Unnamed: 0'])#df = df.drop(columns=['Unnamed: 0'])\n",
    "df_deduplicated  = df.drop_duplicates(subset=['date','expanded_domain']).reset_index(drop=True)\n",
    "merged_df = pd.merge(df, domain_ratings, on='expanded_domain', how='inner')\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "merged_df = pd.merge(merged_df, newsguard, on='expanded_domain', how='inner')\n",
    "\n",
    "merged_df['count'] = merged_df['count'].astype(int) #df['count'] = df['count'].astype(int)\n",
    "merged_df = merged_df.drop_duplicates(subset=['date','expanded_domain']).reset_index(drop=True)\n",
    "# Select specific columns\n",
    "merged_df = merged_df[['date', 'expanded_domain', 'count', 'pc1','Score']]\n",
    "\n",
    "top =['youtube.com']\n",
    "# Step 2: Filter the DataFrame to exclude rows with these domains\n",
    "merged_df = merged_df[~merged_df['expanded_domain'].isin(top)]\n",
    "\n",
    "\n",
    "number_of_share_count = merged_df ['count'].sum()\n",
    "\n",
    "import pandas as pd\n",
    "panel = pd.read_csv('/home/ozturan/covid-conspiracy-narratives/src/panel_daily_urls_withbackfilled_andkenny.csv')\n",
    "panel = panel.drop('Unnamed: 0', axis=1)\n",
    "panel_deduplicated  = panel.drop_duplicates(subset=['id']).reset_index(drop=True)\n",
    "panel_merged_df = panel_deduplicated[(panel_deduplicated['date'] >= '2022-01-01') & (panel_deduplicated['date'] < '2023-05-01') ].reset_index(drop=True)\n",
    "\n",
    "panel_merged_df=panel_merged_df.rename(columns={'domain': 'expanded_domain'})\n",
    "\n",
    "# Ensure 'date' is datetime type\n",
    "panel_merged_df['date'] = pd.to_datetime(panel_merged_df['date'])\n",
    "panel_merged_df = panel_merged_df.groupby(['date', 'expanded_domain', 'pc1']).size().reset_index(name='count')\n",
    "panel_merged_df['count'] = panel_merged_df['count'].astype(int)\n",
    "\n",
    "\n",
    "panel_merged_df = pd.merge(panel_merged_df, newsguard, on='expanded_domain', how='inner')\n",
    "\n",
    "top =['youtube.com','yt.vu']\n",
    "p_merged_df = panel_merged_df[~panel_merged_df['expanded_domain'].isin(top)]\n",
    "number_of_share_count = p_merged_df['count'].sum()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "08d5c7df-ef2b-493f-bad1-d6a7d57504f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Convert date columns to datetime (if not already)\n",
    "p_merged_df['date'] = pd.to_datetime(p_merged_df['date'])\n",
    "merged_df['date'] = pd.to_datetime(merged_df['date'])\n",
    "\n",
    "# Define quintile thresholds\n",
    "quintile_thresholds = [0, 60]\n",
    "\n",
    "# Number of bootstrap samples\n",
    "n_bootstrap = 10\n",
    "\n",
    "# Initialize lists for storing bootstrapped results\n",
    "y_panel_bootstrap = []\n",
    "y_panel_vol_bootstrap = []\n",
    "y_deca_bootstrap = []\n",
    "y_deca_vol_bootstrap = []\n",
    "\n",
    "# Function for bootstrapping with confidence intervals\n",
    "def bootstrap_samples(data, n_bootstrap=n_bootstrap):\n",
    "    return [np.random.choice(data, size=len(data), replace=True) for _ in range(n_bootstrap)]\n",
    "\n",
    "# Function for bootstrapping\n",
    "def bootstrap_ci(data, n_bootstrap=n_bootstrap, ci=95):\n",
    "    if len(data) == 0:\n",
    "        return None  # Return None if data is empty\n",
    "    boot_means = [np.mean(sample) for sample in bootstrap_samples(data, n_bootstrap)]\n",
    "    lower_bound = np.percentile(boot_means, (100 - ci) / 2)\n",
    "    upper_bound = np.percentile(boot_means, 100 - (100 - ci) / 2)\n",
    "    return boot_means, lower_bound, upper_bound\n",
    "\n",
    "# Bootstrapping for **Panel Data** (p_merged_df)\n",
    "for thrshold in quintile_thresholds:\n",
    "    filtered_df = p_merged_df[(p_merged_df['Score'] > thrshold) & (p_merged_df['Score'] <= thrshold + 60)]\n",
    "    \n",
    "    daily_sum = filtered_df.groupby('date')['count'].sum()\n",
    "    daily_sum_all = p_merged_df.groupby('date')['count'].sum()\n",
    "    daily_ratio = daily_sum / daily_sum_all\n",
    "\n",
    "    # Skip if not enough data\n",
    "    if len(daily_sum) < 299:\n",
    "        continue\n",
    "\n",
    "    change_bootstrap = []\n",
    "    change_r_bootstrap = []\n",
    "\n",
    "    for _ in range(n_bootstrap):\n",
    "        sampled_indices_bef = np.random.choice(daily_sum.index[:299], size=299, replace=True)\n",
    "        sampled_indices_aft = np.random.choice(daily_sum.index[299:485], size=(485 - 299), replace=True)\n",
    "\n",
    "        if len(sampled_indices_bef) == 0 or len(sampled_indices_aft) == 0:\n",
    "            continue  # Skip empty cases\n",
    "\n",
    "        sampled_daily_sum_bef = daily_sum.loc[sampled_indices_bef]\n",
    "        sampled_daily_sum_aft = daily_sum.loc[sampled_indices_aft]\n",
    "\n",
    "        sampled_daily_ratio_bef = daily_ratio.loc[sampled_indices_bef]\n",
    "        sampled_daily_ratio_aft = daily_ratio.loc[sampled_indices_aft]\n",
    "\n",
    "        bef_mean = sampled_daily_sum_bef.mean()\n",
    "        aft_mean = sampled_daily_sum_aft.mean()\n",
    "        change = (aft_mean - bef_mean) / bef_mean if bef_mean != 0 else np.nan\n",
    "\n",
    "        bef_ratio_mean = sampled_daily_ratio_bef.mean()\n",
    "        aft_ratio_mean = sampled_daily_ratio_aft.mean()\n",
    "        change_r = (aft_ratio_mean - bef_ratio_mean) / bef_ratio_mean if bef_ratio_mean != 0 else np.nan\n",
    "\n",
    "        change_bootstrap.append(change * 100)\n",
    "        change_r_bootstrap.append(change_r * 100)\n",
    "\n",
    "    if change_bootstrap and change_r_bootstrap:\n",
    "        y_panel_bootstrap.append(bootstrap_ci(change_r_bootstrap))\n",
    "        y_panel_vol_bootstrap.append(bootstrap_ci(change_bootstrap))\n",
    "\n",
    "# Bootstrapping for **Decahose Data** (merged_df)\n",
    "for thrshold in quintile_thresholds:\n",
    "    filtered_df = merged_df[(merged_df['Score'] > thrshold) & (merged_df['Score'] <= thrshold + 60)]\n",
    "    \n",
    "    daily_sum = filtered_df.groupby('date')['count'].sum()\n",
    "    daily_sum_all = merged_df.groupby('date')['count'].sum()\n",
    "    daily_ratio = daily_sum / daily_sum_all\n",
    "\n",
    "    # Skip if not enough data\n",
    "    if len(daily_sum) < 299:\n",
    "        continue\n",
    "\n",
    "    change_bootstrap = []\n",
    "    change_r_bootstrap = []\n",
    "\n",
    "    for _ in range(n_bootstrap):\n",
    "        sampled_indices_bef = np.random.choice(daily_sum.index[:299], size=299, replace=True)\n",
    "        sampled_indices_aft = np.random.choice(daily_sum.index[299:485], size=(485 - 299), replace=True)\n",
    "\n",
    "        if len(sampled_indices_bef) == 0 or len(sampled_indices_aft) == 0:\n",
    "            continue\n",
    "\n",
    "        sampled_daily_sum_bef = daily_sum.loc[sampled_indices_bef]\n",
    "        sampled_daily_sum_aft = daily_sum.loc[sampled_indices_aft]\n",
    "\n",
    "        sampled_daily_ratio_bef = daily_ratio.loc[sampled_indices_bef]\n",
    "        sampled_daily_ratio_aft = daily_ratio.loc[sampled_indices_aft]\n",
    "\n",
    "        bef_mean = sampled_daily_sum_bef.mean()\n",
    "        aft_mean = sampled_daily_sum_aft.mean()\n",
    "        change = (aft_mean - bef_mean) / bef_mean if bef_mean != 0 else np.nan\n",
    "\n",
    "        bef_ratio_mean = sampled_daily_ratio_bef.mean()\n",
    "        aft_ratio_mean = sampled_daily_ratio_aft.mean()\n",
    "        change_r = (aft_ratio_mean - bef_ratio_mean) / bef_ratio_mean if bef_ratio_mean != 0 else np.nan\n",
    "\n",
    "        change_bootstrap.append(change * 100)\n",
    "        change_r_bootstrap.append(change_r * 100)\n",
    "\n",
    "    if change_bootstrap and change_r_bootstrap:\n",
    "        y_deca_bootstrap.append(bootstrap_ci(change_r_bootstrap))\n",
    "        y_deca_vol_bootstrap.append(bootstrap_ci(change_bootstrap))\n",
    "\n",
    "# Extract means and confidence intervals\n",
    "def extract_ci(bootstrap_results):\n",
    "    return (\n",
    "        [b[0] for b in bootstrap_results if b is not None],  # Means\n",
    "        [b[1] for b in bootstrap_results if b is not None],  # Lower CI\n",
    "        [b[2] for b in bootstrap_results if b is not None]   # Upper CI\n",
    "    )\n",
    "\n",
    "# Extract values for both datasets\n",
    "y_panel_means, _, _ = extract_ci(y_panel_bootstrap)\n",
    "y_panel_vol_means, _, _ = extract_ci(y_panel_vol_bootstrap)\n",
    "\n",
    "y_deca_means, _, _ = extract_ci(y_deca_bootstrap)\n",
    "y_deca_vol_means, _, _ = extract_ci(y_deca_vol_bootstrap)\n",
    "\n",
    "# Categories for plotting\n",
    "quintile_labels = [\"Low\", \"High\"]\n",
    "colors_panel = ['darkorange', 'orange', 'gold']\n",
    "colors_decahose = ['darkblue', 'mediumblue', 'royalblue']\n",
    "\n",
    "def create_boxplot(ax, title, data, colors, row):\n",
    "    if len(data) == 0:\n",
    "        ax.set_title(title + \" (No Data)\", fontsize=14)\n",
    "        return\n",
    "\n",
    "    box = ax.boxplot(data, patch_artist=True, medianprops=dict(color='black'), widths=0.5, whis=4,showfliers=False)\n",
    "\n",
    "    ax.set_title(title, fontsize=14)\n",
    "    ax.set_ylabel('Percentage Change', fontsize=12)\n",
    "    ax.set_xlabel('Quintile Ranges', fontsize=12)\n",
    "    ax.set_xticks(range(1, len(data) + 1))\n",
    "    ax.set_xticklabels(quintile_labels[:len(data)], rotation=0, ha='right', fontsize=11)\n",
    "    ax.grid(axis='y', linestyle='--', alpha=0.7)\n",
    "    ax.axhline(0, color='black', linewidth=2)\n",
    "\n",
    "    # **Set different y-axis limits based on the row**\n",
    "    if row == 0:\n",
    "        ax.set_ylim(-30, 5)  # First row (Decahose)\n",
    "    else:\n",
    "        ax.set_ylim(-5, 20)  # Second row (Panel)\n",
    "\n",
    "    # Color each box according to dataset\n",
    "    for patch, color in zip(box['boxes'], colors):\n",
    "        patch.set_facecolor(color)\n",
    "\n",
    "# Create boxplots\n",
    "fig, axs = plt.subplots(2, 2, figsize=(8, 10))\n",
    "\n",
    "\n",
    "for i in range(2):  # Loop through rows\n",
    "    for j in range(2):  # Loop through columns\n",
    "        row = i  # Store the row index\n",
    "        if i == 0 and j == 0:\n",
    "            create_boxplot(axs[i, j], 'Decahose: Volume Change (%)', y_deca_vol_means, colors_decahose, row)\n",
    "        elif i == 0 and j == 1:\n",
    "            create_boxplot(axs[i, j], 'Panel: Volume Change(%)', y_panel_vol_means, colors_panel, row)\n",
    "            \n",
    "            \n",
    "        elif i == 1 and j == 0:\n",
    "            create_boxplot(axs[i, j], 'Decahose: Market Share Change (%)', y_deca_means, colors_decahose, row)\n",
    "\n",
    "            \n",
    "            \n",
    "        elif i == 1 and j == 1:\n",
    "            create_boxplot(axs[i, j], 'Panel: Market Share Change (%)', y_panel_means, colors_panel, row)\n",
    "            \n",
    "\n",
    "plt.tight_layout()\n",
    "plt.savefig('/home/ozturan/covid-conspiracy-narratives/src/musk/submission/figures/fig3_bootstrapped_ci.png', dpi=600, bbox_inches='tight', pad_inches=0.1)\n",
    "plt.savefig('/home/ozturan/covid-conspiracy-narratives/src/musk/submission/figures/fig3_bootstrapped_ci.pdf', dpi=600, bbox_inches='tight', pad_inches=0.1, format='pdf')\n",
    "\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "3c0f3114-b035-4697-a98a-3d38066ddbac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[np.float64(1.0292911162843787),\n",
       "  np.float64(2.3579508957109816),\n",
       "  np.float64(1.9021864636608838),\n",
       "  np.float64(1.4514436150690202),\n",
       "  np.float64(2.9698880888882053),\n",
       "  np.float64(0.9253946096906815),\n",
       "  np.float64(0.9978922507084291),\n",
       "  np.float64(0.8967103833963155),\n",
       "  np.float64(1.734539052370718),\n",
       "  np.float64(2.2174860065106654)],\n",
       " [np.float64(-16.472884382311932),\n",
       "  np.float64(-16.56827916229264),\n",
       "  np.float64(-16.14732485690258),\n",
       "  np.float64(-17.498880567092552),\n",
       "  np.float64(-16.11771231659607),\n",
       "  np.float64(-16.440848847562982),\n",
       "  np.float64(-16.755958386065277),\n",
       "  np.float64(-16.832117527111137),\n",
       "  np.float64(-16.17374942618558),\n",
       "  np.float64(-15.973001361148295)]]"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_deca_vol_means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "9e9ee22b-baf8-45e6-b30b-5137c3e3822d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[np.float64(18.557481190609064),\n",
       "  np.float64(18.37808595833103),\n",
       "  np.float64(18.44546767652729),\n",
       "  np.float64(18.42898486429174),\n",
       "  np.float64(18.630141265977056),\n",
       "  np.float64(18.455437220588227),\n",
       "  np.float64(18.82948269851316),\n",
       "  np.float64(18.518693656341572),\n",
       "  np.float64(18.59540005251643),\n",
       "  np.float64(18.48981965685273)],\n",
       " [np.float64(-1.9419147582897676),\n",
       "  np.float64(-1.9014697567255663),\n",
       "  np.float64(-2.0494405103420825),\n",
       "  np.float64(-1.9151871836893963),\n",
       "  np.float64(-1.9431978072083247),\n",
       "  np.float64(-1.9266335526360983),\n",
       "  np.float64(-2.064268923052276),\n",
       "  np.float64(-1.9944917147671681),\n",
       "  np.float64(-1.9364688223315962),\n",
       "  np.float64(-1.9203156101135974)]]"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "\n",
    "y_deca_means\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "3c1a5005-415e-4297-ab71-186184c7a157",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[np.float64(-23.75478000015378),\n",
       "  np.float64(-22.483394352984142),\n",
       "  np.float64(-23.492097032248296),\n",
       "  np.float64(-23.93117674230499),\n",
       "  np.float64(-24.250708679868296),\n",
       "  np.float64(-23.59134049122279),\n",
       "  np.float64(-23.456621128738238),\n",
       "  np.float64(-23.495135259092276),\n",
       "  np.float64(-23.33334744828346),\n",
       "  np.float64(-23.01793598597706)],\n",
       " [np.float64(-26.89944299104412),\n",
       "  np.float64(-26.332170183708975),\n",
       "  np.float64(-26.31765380347889),\n",
       "  np.float64(-25.531523778705186),\n",
       "  np.float64(-26.523039434809895),\n",
       "  np.float64(-25.809395885932624),\n",
       "  np.float64(-26.697822608350332),\n",
       "  np.float64(-25.78524538294632),\n",
       "  np.float64(-26.515575612142868),\n",
       "  np.float64(-26.388877479761664)]]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_panel_vol_means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "aa3d2c15-271c-461f-9100-c1883e37782b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[np.float64(5.809495842617402),\n",
       "  np.float64(5.8239871471016595),\n",
       "  np.float64(5.119134201441073),\n",
       "  np.float64(5.1981353419286425),\n",
       "  np.float64(5.573828996716175),\n",
       "  np.float64(5.758413329095002),\n",
       "  np.float64(5.552107037598027),\n",
       "  np.float64(5.784112286588817),\n",
       "  np.float64(5.326056688155817),\n",
       "  np.float64(5.989410085084559)],\n",
       " [np.float64(-0.8549013492280715),\n",
       "  np.float64(-0.823446920152216),\n",
       "  np.float64(-0.9234214957580503),\n",
       "  np.float64(-0.9515463656634088),\n",
       "  np.float64(-0.8495603994390419),\n",
       "  np.float64(-0.9087572000859842),\n",
       "  np.float64(-0.8653983832349595),\n",
       "  np.float64(-0.8872195526703205),\n",
       "  np.float64(-0.8615501586170125),\n",
       "  np.float64(-0.8657376146935439)]]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_panel_means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a899be05-9fe1-44d4-b1c1-37ed916df293",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
